Evaluation of creel survey methodology for steelhead fisheries on the Quillayute and Hoh Rivers

The Hoh and Quillayute river systems provide two of the most popular steelhead fisheries in Washington State with the majority of fish caught being wild steelhead. Beginning in the mid-1970s in the Quillayute River and the 2000s in the Hoh River, the Washington Department of Fish and Wildlife (WDFW) has conducted on-site steelhead creel surveys (i.e., ‘coastal creels') in order to estimate fishing pressure and catch of winter steelhead. In both rivers, fishery regulations for steelhead have changed over time as have the specific goals of the coastal creel surveys and methods used to generate estimates of catch. The purpose of this report is to summarize the present day goals and methods of the coastal creel surveys, evaluate whether these goals are being met based on current methodology, and propose recommendations for future surveys.

The primary goal of the coastal creel surveys is to generate unbiased estimates of total end-of-season recreational steelhead catch, where catch is a combination of harvest and catch and release (C&R). During an on-site creel survey, fishing pressure (angler effort) and angler success (catch per unit effort - CPUE) data are collected via anger counts and interviews, respectively. These data are subsequently used to make expanded estimates of effort and catch. For unbiased estimates to be generated, creel survey data must be collected in a spatially and temporally representative manner. This assumption may not be met with the current methodology resulting in an underestimate of catch. I make several recommendations to improve the spatial (e.g., effort expansion surveys) and temporal (e.g., randomized sample schedule, use of day length in effort calculation) coverage.

A secondary goal of the coastal creel surveys is to produce catch estimates with a precision of ± 20% at the 95% confidence level, which corresponds to a coefficient of variation (CV) of 10%. A method for calculating precision is provided based on previously derived equations outlined in Pollock et al. (1994) and Hahn et al. (2000). Catch and precision estimates were generated for each pairwise combination of fish origin (wild, hatchery) and fate (harvested, released) in the Upper and Lower river-sections of the Hoh River during the 2014-15 creel survey season. The resulting CVs ranged from 11% to 25%, which are all greater than the specified goal of 10% but relatively precise based on standards for other estimates such as monitoring of ESA-listed species (Crawford and Rumsey 2011). Before additional time and funds are invested in lowering the precision of the creel estimate, the current precision goal should be examined with regards to how the resulting data will be used for management and decision making.